We’re excited to announce that Sensu 2.0 Beta is now publicly available for testing! Sensu 2.0 is easier to install and operate, requiring only a single sensu-backend
process and sensu-agent
process that communicate directly with each other. Our new API-driven approach gives you more flexibility when integrating Sensu into your existing infrastructure and tools. We believe this newest release is built to scale and meet the needs not only of your infrastructure today, but for years to come.
How I feel right now
What’s new
We have worked incredibly hard to make the use of Sensu easier for today and the long run, and that comes with some cost to backwards compatibility. The Sensu 2.0 API differs from Sensu 1.0 in both how Sensu is configured and the data models used to represent the state of your systems. Metrics always felt bolted on in Sensu 1.0 — we’ve addressed that with our new data model, which makes metrics a first-class citizen of the Sensu ecosystem. We also introduced multiple API endpoints for managing your checks, handlers, and mutators. As you build systems that interact with and manage Sensu 2.0, we know that the reliability and stability of the API is of utmost importance; we wanted to ensure a stable API once the beta launched, which gives us flexibility during the closed and open alpha stages to address feedback.
We also wanted to make sure the new product was able to support even some of the largest installations. Prior to Sensu 2.0, larger installations may have required multiple Sensu clusters. In our goal of decreasing operator burden, we wanted to make sure we addressed this. Now, a single instance of the Sensu 2.0 backend is capable of supporting more than 10,000 requests per second, with as many or more simultaneously connected agents. Our goal is to have even some of the largest installations of Sensu today (with tens of thousands of Sensu clients) run on as few Sensu backend processes as possible.
Documentation + feedback FTW
While launching the new Sensu documentation website, engineering was also preparing documentation for Sensu 2.0, which includes instructions on getting started with Sensu 2.0, reference material, and a number of guides that will help you discover the right features to solve your monitoring challenges. The documentation website is itself open source, and we encourage you to open pull requests or submit issues if you come across inaccurate or incorrect documentation.
Since open sourcing the Sensu 2.0 project in February, the engineering team has been focusing on production readiness. For the past month, we’ve been polishing and completing our documentation, establishing API stability, and taking performance into account. We’re excited to get more feedback from you on how we can improve the onboarding experience.
As you explore the project and documentation, we hope you’ll join us in our community Slack. There are a number of developers and community members in the #sensu2 channel that are available to answer questions about the project and help you get started. We can’t wait to hear what you think!
Thanks to Anna MacLachlan and Sensu.